Please use this identifier to cite or link to this item: https://repositori.uma.ac.id/handle/123456789/25339
Title: Klasifikasi Status Gizi Balita Menggunakan Algoritma Support Vector Machine (SVM) (Studi Kasus: Puskesmas Kecamatan Gunung Meriah)
Other Titles: Classification of Toddler Nutritional Status Using the Support Vector Machine (SVM) Algorithm (Case Study: Gunung Meriah District Health Center)
Authors: Firanti, Mira
metadata.dc.contributor.advisor: Muliono, Rizki
Keywords: gizi balita;SVM;klasifikasi;akurasi;toddler nutrition;clasification;accuracy
Issue Date: 1-Apr-2024
Publisher: UNIVERSITAS MEDAN AREA
Series/Report no.: NPM;198160017
Abstract: Kasus ketidakseimbangan gizi merupakan masalah gizi paling umum di Indonesia, terutama pada anak usia 0-5 tahun yang rentan terhadap kekurangan gizi. Kekurangan gizi kronis berdampak signifikan terhadap pertumbuhan, kemampuan kognitif, dan kesehatan mental anak, serta meningkatkan risiko kematian. Kabupaten Aceh Singkil memiliki angka prevalensi stunting tinggi, khususnya di Kecamatan Gunung Meriah yang mencatat angka stunting tertinggi sebesar 3,7%. Metode penilaian gizi yang masih manual seringkali menghasilkan data yang kurang akurat. Penelitian ini bertujuan mengembangkan sistem klasifikasi status gizi balita menggunakan algoritma Support Vector Machine (SVM) berbasis web untuk membantu petugas gizi dalam menentukan status gizi balita dengan lebih akurat. Metode penelitian menggunakan model Waterfall dan pemodelan UML, dengan data antropometri balita yang dibagi dalam rasio 80:20 untuk pelatihan dan pengujian. Hasil menunjukkan bahwa kernel linear SVM mencapai akurasi 86%. Analisis lanjutan menggunakan kernel polynomial dan RBF meningkatkan akurasi menjadi 88,75%, dengan presisi 89,70%, recall 88,75%, dan F1-score 88,69%. Algoritma SVM efektif dalam mengklasifikasikan status gizi balita, membantu petugas gizi dalam membuat keputusan yang lebih akurat. Penelitian ini menunjukkan bahwa SVM dapat meningkatkan akurasi penilaian status gizi balita, mendukung upaya peningkatan kesehatan anak di Aceh Singkil. Cases of nutritional imbalance are the most common nutritional problem in Indonesia, especially in children aged 0-5 years who are vulnerable to malnutrition. Chronic malnutrition has a significant impact on children's growth, cognitive abilities and mental health, and increases the risk of death. Aceh Singkil Regency has a high stunting prevalence rate, especially in Gunung Meriah District which recorded the highest stunting rate at 3.7%. Manual nutritional assessment methods often produce inaccurate data. This research aims to develop a classification system for the nutritional status of toddlers using the web-based Support Vector Machine (SVM) algorithm to assist nutrition officers in determining the nutritional status of toddlers more accurately. The research method uses the Waterfall model and UML modeling, with toddler anthropometric data divided in a ratio of 80:20 for training and testing. The results show that the linear SVM kernel achieves 86% accuracy. Advanced analysis using polynomial kernels and RBF increased accuracy to 88.75%, with precision of 89.70%, recall of 88.75%, and F1-score of 88.69%. The SVM algorithm is effective in classifying the nutritional status of toddlers, helping nutrition officers make more accurate decisions. This research shows that SVM can increase the accuracy of assessing the nutritional status of toddlers, supporting efforts to improve children's health in Aceh Singkil.
Description: 51 Halaman
URI: https://repositori.uma.ac.id/handle/123456789/25339
Appears in Collections:SP - Informatic Engineering

Files in This Item:
File Description SizeFormat 
198160017 - Mira Firanti - Fulltext.pdfCover, Abstract, Chapter I, II, III, V, Bibliography1.87 MBAdobe PDFView/Open
198160017 - Mira Firanti - Chapter IV.pdf
  Restricted Access
Chapter IV402.16 kBAdobe PDFView/Open Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.